Network elements such as a Serving GPRS Support Node (SGSN) and a Gateway GPRS Support Node (GGSN) of the GPRS usually send to each other echo request messages so as to poll that the other GSN, i.e. SGSN or GGSN, is alive.
A GSN receiving an echo request message responds with an echo response message which contains a restart counter value as a parameter. The GSN which receives the echo response message compares the received restart counter value with a previous restart counter value stored for that particular peer GSN.
If the previously stored restart counter value differs from the restart counter value received in the echo response message, the GSN that sent the echo response message is considered by the GSN that received the echo response message as being restarted. The received new restart counter value is stored by the receiving GSN, replacing the value previously stored for the sending GSN. Subsequently, the GSN that received the echo response message with the new (unexpected) restart counter value considers all subscriber contexts, i.e. Packet Data Protocol contexts (PDP contexts), relating to the sending GSN as inactive.
A VLR (Visitor Location Register) restart procedure is a similar procedure, wherein the VLR informs a HLR (Home Location Register) of the restart. When the VLR has been restarted, the complete set of subscriber information relevant to the VLR is transmitted to the VLR.
FIG. 1 shows a principle diagram indicating information transfers and processings of such an echoing procedure performed between two network elements NE1 and NE2. In FIG. 1, the processing starts at the top and moves to the bottom.
Initially, the network element NE2 transmits an echo request message to the network element NE1 which responses with an echo request response message including its restart counter value (N=n). If the restart counter values of both network elements are equal, the NE2 assumes that no restart has been performed since the last echo request.
Thereafter, mobile stations (MS) of the mobile communication network activate new subscriber contexts to the network elements, such that new contexts are created in the NE1 and NE2. The NE1 is then restarted due to a failure or the like, and its restart counter is incremented. Subsequently, MSs activate further subscriber contexts which are created in the NE1 and NE2.
If the NE2 now transmits an echo request message to the NE1, the restart counter value received in the echo response message differs from the value previously stored for the NE1. Accordingly, the NE2 assumes a restart of the NE1 and inactivates all subscriber contexts of the NE1, although some subscriber contexts were created after the restart of the NE1 and are thus still valid.
Thus, unnecessary subscriber context re-activations have to be performed and the downtime of the service is high.